The experiments reported in this paper indicate that a pure culture of a virulent Type I pneumococcus as transferred under ordinary conditions is made up of individuals which are not identical in virulence for animals,—in bile solubility, or in their power to multiply under unfavorable conditions.
By mechanical analysis alone no avirulent strains were obtained from the virulent strain, but one, Strain C, was separated which proved to be much more virulent for rabbits than its fellows.
The virulent, colony-pure, Type I pneumococcus was allowed to multiply for several transplants in a series of cultures in (a) pneumococcus antiserum broth, (b) bile broth, and (c) slightly acid broth. When cultures from each series were plated on blood agar there appeared amid the ordinary pneumococcus colonies some that were totally different therefrom. Cultures from these new colonies were macroscopically bile-soluble, fermented inulin, and produced methemoglobin. Their occurrence under these different sets of conditions as well as careful controls excluded the possibility of a derivation by contamination.
The new strains were avirulent and did not become virulent on repeated passage through mice. Antisera procured by the injection of them into rabbits were found to contain agglutinins for Type I pneumococci, but no protective antibodies.
The avirulent strains are evidently pneumococci. It is possible that they may have been present in the parent strain and the unfavorable conditions of growth (immune serum, bile, slight acidity) might have allowed them to multiply while repressing the virulent forms. The individuals composing the pneumococcus culture are not all equally bile-soluble. By means of differential bile solubility, the avirulent forms were obtained within 12 hours from the parent strain though only twice in six experiments. It is possible that the avirulent individuals might not have been present in the 26 hour stock culture which yielded one of the positive results and that the organisms surviving the lysis might have retained sufficient bile after washing with isotonic salt solution for this to be responsible for the development of the variant. The fact that the identical experiment with the virulent Derivative Strain C revealed no avirulent forms renders it likely that the majority of the avirulent forms originated from the virulent individuals under the unfavorable conditions. The correctness of this interpretation is proven by the finding of these avirulent forms in virulent single cell derivative strains grown under unfavorable conditions. By contrast, strains of these avirulent forms descended from single diplococci did not revert to virulence. Apparently there had occurred a genuine bacterial mutation. These results in general are in accord with those of Griffith8 who has made a more careful study of the properties of variants of pneumococci. He has recorded the occasional reversion of a variant when the single colony culture has been made after one serum passage. There is no reference in his article to any experiments with strains derived by the isolation of single diplococci.
Further evidence of the composite nature of the parent strain as ordinarily grown is presented by the difference in the effect of long continued growth in immune serum on the pure line strains and on the parent strain itself. The pure line strains, equally as virulent for mice as the parent, and one of them, Strain C, more virulent for rabbits, became less virulent than the parent strain after twelve transplants in immune serum. Growth in normal horse serum did not affect the virulence. The single cell derivative, moreover, was found to be the less resistant to bile when grown in increasing concentrations of this lytic substance. The virulent Derivative Strain C became avirulent in eleven transplants in acid broth pH 6.8 while the parent strain, though reduced in virulence, was still lethal in doses of 0.001 cc. after the same number of transplants in this medium.